[PATCH] D91038: [LoopIdiom] Introduce 'left-shift until bittest' idiom

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri created this revision.
lebedev.ri added reviewers: fhahn, craig.topper, ychen, spatel, ovmold, efriedma, evstupac, jdoerfert.
lebedev.ri added a project: LLVM.
Herald added a subscriber: hiraditya.
lebedev.ri requested review of this revision.

The motivation here is the following inner loop in fp16/fp24 -> fp32 expander,
that runs as part of the floating-point DNG decompression in RawSpeed library:
https://github.com/darktable-org/rawspeed/blob/cd380bb9a209bd2e7a0e7022b0cab04528d151e7/src/librawspeed/decompressors/DeflateDecompressor.cpp#L112-L115

  while (!(fp32_fraction & (1 << 23))) {
    fp32_exponent -= 1;
    fp32_fraction <<= 1;
  }

As one might notice, that loop is currently uncountable, and that whole code stays scalar.
Yet, it is rather trivial to make that loop countable:

  https://godbolt.org/z/do8WMz

and we can prove that via alive2:

  https://alive2.llvm.org/ce/z/7vQnji (ha nice, isn't it?)

... and that allow for the whole fp16->fp32 code to vectorize:

  https://godbolt.org/z/7hYr13

Now, while i'd love to get there, i feel like i should take it in steps.

For now, this introduces support for the most basic case,
where the bit position is known as a variable,
and the loop *will* go away (has no live-outs other than the recurrence,
no extra instructions in the loop).

I have added sufficient (i believe) test coverage,
and alive2 is happy with those transforms.

Thoughts?


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D91038

Files:
  llvm/lib/Transforms/Scalar/LoopIdiomRecognize.cpp
  llvm/test/Transforms/LoopIdiom/X86/left-shift-until-bittest.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D91038.303730.patch
Type: text/x-patch
Size: 43235 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201108/fcc17d73/attachment.bin>